Top all-time · #4994
Blog on Machine, Think!
Personal blog · machinethink.net/blogaihardware
Blog on Machine, Think! is a personal blog at machinethink.net about ai and hardware. Since 2017, 15 different Hacker News readers have linked it 17 times, most often “One-shot object detection”. The last post was in 2022, but the archive is still widely read.
Most-linked posts?
- One-shot object detection3 people · 3 mentions · first linked 2018-06-09
- A peek inside Core ML2 people · 2 mentions · first linked 2017-09-03
- The lost art of 3D rendering without shaders2 people · 2 mentions · first linked 2017-01-21
What Hacker News says?
You can do some refinement right now already (from WWDC 2019). But yes, most models are trained on NVIDIA GPUs that are deep learning. All of the other model types can already be trained on regular CPUs. https://machinethink.net/blog/coreml-training-part1/
Comment linking /blog/coreml-training-part1/
on “Apple announces it will switch to its own processors for future Macs”Submitted The lost art of 3D rendering without shaders as a story · 319 points
HN discussion- nl · 2021-07-16
It's not misleading assuming you know the field. If you don't you might not realize they are comparing ResNet (designed for ultimate performance) vs EffcientNet to show how close it gets in accuracy in the FLOPs budget. Note that the best accuracy for EffcientNet is roughly 1-2% better than ResNet/ResNeXT/SENet/etc but does have a much better FLOPs budget. But these other architectures were never optimised for FLOPs…
Comment linking /blog/mobile-architectures/
on “Alphafold” Submitted Apple machine learning in 2020: What’s new? as a story · 95 points
HN discussionIs it? The edge TPU can do MobileNet V2 at 100 FPS. An iPhone 7 can do it at 145 FPS (source https://machinethink.net/blog/mobilenet-v2/ ) The deal is though that the Edge TPU is able to do it at much lower power.
Comment linking /blog/mobilenet-v2/
on “Google Coral Edge TPU”Submitted Machine learning on mobile: on the device or in the cloud? as a story · 81 points
HN discussionSubmitted The lost art of 3D rendering without shaders as a story · 4 points
HN discussionSubmitted A peek inside Core ML as a story · 3 points
HN discussionSubmitted Convolutional neural networks on the iPhone with VGGNet as a story · 3 points
HN discussionSubmitted One-shot object detection as a story · 2 points
HN discussionSubmitted One-shot object detection as a story · 2 points
HN discussionSubmitted Compressing deep neural nets as a story · 2 points
HN discussionSubmitted New mobile neural network architectures benchmarked in iOS as a story · 1 points
HN discussionSubmitted One-shot object detection as a story · 1 points
HN discussionSubmitted A peek inside Core ML as a story · 1 points
HN discussionSubmitted YOLO: Core ML versus MPSNNGraph as a story · 1 points
HN discussionSubmitted YOLO: Core ML versus MPSNNGraph as a story · 1 points
HN discussion
